Pathological correlates of late-onset dementia in a multicentre, community-based population in England and Wales

Background We have undertaken a large unselected, community-based neuropath ology study in an elderly (70-103 years) UK population in relation to prosp ectively evaluated dementia status. The study tests the assumption that dem enting disorders as defined by current diagnostic protocols underlie this s yndrome in the community at large. Methods Respondents in the Medical Research Council Cognitive Function and Ageing Study were approached for consent to examine the brain at necropsy. Dementia status was assigned by use of the automated geriatric examination for computer-assisted taxonomy algorithm. Neuropathological features were s tandardised by use of the protocol of the Consortium to Establish a Registr y of Alzheimer's Disease, which assesses the severity and distribution of A lzheimer-type pathology, vascular lesions, and other potential causes of de mentia. A statistical model of dementia risk related predominantly to Alzhe imer-type and vascular pathology was developed by multivariate logistic reg ression. Findings We report on the first 209 individuals who have come to necropsy. The median age at death was 85 years for men, and 86 years for women. Cereb rovascular (78%) and Alzheimer-type (70%) pathology were common. Dementia w as present in 100 (48%), of whom 64% had features indicating probable or de finite Alzheimer's disease. However, 33% of the 109 non-demented people had equivalent densities of neocortical neuritic plaques. Some degree of neoco rtical neurofibrillary pathology was found in 61% of demented and 34% of no n-demented individuals. Vascular lesions were equally common in both groups , although the proportion with multiple vascular pathology was higher in th e demented group (46% vs 33%). Interpretation Alzheimer-type and Vascular pathology were the major patholo gical correlates of cognitive decline in this elderly sample, as expected, but most patients had mixed disease. There were no clear thresholds of thes e features that predicted dementia status. The findings therefore challenge conventional dementia diagnostic criteria in this setting. Additional fact ors must determine whether moderate burdens of cerebral Alzheimer-type path ology and vascular lesions are associated with cognitive failure.